change to leap-lab backend. Support feedback interval. Reduce cocurrent lags.

unilabos/utils/tools.py

@@ -17,6 +17,14 @@ try:
             default=json_default,
         )
 
+    def fast_loads(data) -> dict:
+        """JSON 反序列化，优先使用 orjson。接受 str / bytes。"""
+        return orjson.loads(data)
+
+    def fast_dumps_str(obj, **kwargs) -> str:
+        """JSON 序列化为 str，优先使用 orjson。用于需要 str 而非 bytes 的场景（如 ROS msg）。"""
+        return orjson.dumps(obj, option=orjson.OPT_NON_STR_KEYS, default=json_default).decode("utf-8")
+
     def normalize_json(info: dict) -> dict:
         """经 JSON 序列化/反序列化一轮来清理非标准类型。"""
         return orjson.loads(orjson.dumps(info, default=json_default))
@@ -29,6 +37,14 @@ except ImportError:
     def fast_dumps_pretty(obj, **kwargs) -> bytes:  # type: ignore[misc]
         return json.dumps(obj, indent=2, ensure_ascii=False, cls=TypeEncoder).encode("utf-8")
 
+    def fast_loads(data) -> dict:  # type: ignore[misc]
+        if isinstance(data, bytes):
+            data = data.decode("utf-8")
+        return json.loads(data)
+
+    def fast_dumps_str(obj, **kwargs) -> str:  # type: ignore[misc]
+        return json.dumps(obj, ensure_ascii=False, cls=TypeEncoder)
+
     def normalize_json(info: dict) -> dict:  # type: ignore[misc]
         return json.loads(json.dumps(info, ensure_ascii=False, cls=TypeEncoder))